An IIsi in my SE/30? It's more likely than you think!

[luRaichu]

just take the IIsi chipset and squish it onto an SE/30 form factor logic board

 

[mikes-macs]

This is interesting to read. Looking forward to reading more on this.

 

[joshc]

Does this present hope for a cheaper alternative for a Micron Xceed greyscale internal video cards?

I would classify the existing options as cheap compared with what Micron Xceeds were going for.

 

[Bolle]

Wondering if such might be possible for the SE/30 A/B at 512x384 target resolution?

The Apple 12" 512x384 video mode has 128 pixels of blanking added to it for a total line length of 640 pixels.
The compact Mac video signal specifies 192 pixels of blanking so a total line length of 704 pixels is needed. Generating a 512x384 video signal using a total length of 704 pixels will result in a usable picture on a stock analogboard. That's what I am doing on my grayscale card for the bumped up resolution.
Feeding an Apple 12" 512x384 signal with 640 pixels total line length (after mangling the horizontal drive signal to get appropriate pulse durations) to a compact Mac analog board will result in the picture folding over. I wasn't able to shift the active area around by modifying the HSYNC pulse to get rid of the foldover.

As it seems the output resolutions are hardwired in the RBV and can't be configured from software that would prevent the use of 512x384.

 

[Trash80toHP_Mini]

Thanks, unfortunate that. Can an analog video output bodge work, given that limitation in RBV?

How about digging up and posting that pretty IIsi/Compact ratsnest as inspiration for the dreamers?

 

[croissantking]

I would classify the existing options as cheap compared with what Micron Xceeds were going for.

Especially if you opt to build your own neckboard and harness which I am only partway through — but I do have 25ft of 18AWG purple wire ready and waiting!

 

[zigzagjoe]

The Apple 12" 512x384 video mode has 128 pixels of blanking added to it for a total line length of 640 pixels.
The compact Mac video signal specifies 192 pixels of blanking so a total line length of 704 pixels is needed. Generating a 512x384 video signal using a total length of 704 pixels will result in a usable picture on a stock analogboard. That's what I am doing on my grayscale card for the bumped up resolution.
Feeding an Apple 12" 512x384 signal with 640 pixels total line length (after mangling the horizontal drive signal to get appropriate pulse durations) to a compact Mac analog board will result in the picture folding over. I wasn't able to shift the active area around by modifying the HSYNC pulse to get rid of the foldover.

As it seems the output resolutions are hardwired in the RBV and can't be configured from software that would prevent the use of 512x384.

Ah, you're right.

The timings and resolutions are hardwired, yet for some reason they aren't defined in the RBV datasheet. They're broken out in more detail in the IIsi document. I was assuming for some reason that 010 was a modified compact mac video timing first, but that's not the case. Oops.

 

[nottomhanks]

This is insane. I’d love to have a IIsi in an SE/30!!!

 

[David Cook]

FYI: Apple planned to stuff a 16MHz and 25MHz IIci into an SE/30 package. They also planned to replace the IIcx with a 16MHz IIci.

boxMacIIci EQU 5 ; Aurora 25MHz 3 slot package <3.6>
boxAuroraCX16 EQU 8 ; Aurora 16MHz 3 slot package (never shipped) <3.6>
boxAuroraSE25 EQU 9 ; Aurora 25MHz SE30 package (reserved for future) <3.6>
boxAuroraSE16 EQU 10 ; Aurora 16MHz SE30 package (reserved for future) <3.6>

 

[adespoton]

So why did we end up with the bus-limited CC instead of an SE/30ci? Just component pricing?

 

[cheesestraws]

So why did we end up with the bus-limited CC instead of an SE/30ci? Just component pricing?

Because the Colour Classic was not the successor to the SE/30, it was a successor to the Classics. There was no successor to the SE/30, which was a II-class machine in a compact case.

 

[zigzagjoe]

FYI: Apple planned to stuff a 16MHz and 25MHz IIci into an SE/30 package. They also planned to replace the IIcx with a 16MHz IIci.

boxMacIIci EQU 5 ; Aurora 25MHz 3 slot package <3.6>
boxAuroraCX16 EQU 8 ; Aurora 16MHz 3 slot package (never shipped) <3.6>
boxAuroraSE25 EQU 9 ; Aurora 25MHz SE30 package (reserved for future) <3.6>
boxAuroraSE16 EQU 10 ; Aurora 16MHz SE30 package (reserved for future) <3.6>

Yep, noticing these constants being used in the RBV video driver is what led me to thinking about this. Then the RBV docs sealed the deal.

 

[Jockelill]

Aside from the work that'd go into creating a reloaded SE type board, the build itself would be more difficult than most reloaded boards due to the use of QFPs. It would also require a grayscale-type harness, a neckboard, and maybe something to address issues with softpower.... In my experience the EGRET will pout with its arms crossed and not let the system out of reset if you forcibly power the system, but maybe that can be addressed.

Additional difficulties... a hypothetical PDS slot would be 25mhz, so any synchronous or partially synchronous PDS cards may not work as they need to either be fully asynchronous in respect to strobe timings and/or be able to meet internal timings when provided a 25mhz clock. From what I've heard bus-overclocking IIsi results in most PDS cards and the original nubus riser becoming flakey or not working. Nubus is a possibility, but would require a reloaded/bespoke adapter on top of physical fitment troubles.

Cache cards and cache slot upgrades would be on the table electrically but mechanically remains an open question. PPC, it depends on exactly how the PPC ROMs are used. ISTR those cards have a complete PPC ROM that fully takes over which would then undo our changes to get the video activated.

Definitely some interesting possibilities here but it'll never be a cheap or mainstream affair, if it comes to pass.

Wouldn't the PDS be 20Mhz? Since that is what the IIsi stock is running .

Awesome work as always!!